MathDuels - Solver Rating: leaderboard

MathDuels (2026) is a self-play mathematics benchmark: each model authors 30 hardened problems and solves those the other models wrote, with symbolic answer checking and a verifier removing ill-posed problems; ratings come from a Rasch model on an Elo-style scale. This board is the solver rating from solving the problems the other models wrote, in rating points on an open scale; higher is better.

Metric: Solver rating (Elo-style points, open scale): Rasch ability from solving the other models' problems; 19 models each author 30 hardened problems (six mathematical domains) and solve every problem authored by the others; answers checked symbolically, ill-posed problems removed by a GPT-5.4-high verifier; Rasch model on a shared rating scale anchored at Grok-4.1-fast-high = 1500, k = 30; higher is better. Source: arxiv.org. Saturation forecast: Around June 2027. 15 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-5.4High2268#135
2Gemini 3.1 Pro (Preview)High2214#99
3GPT-5.2High2047#185
4Claude Opus 4.6High2043#113
5Qwen 3.5 397B A17B1972#264
6Grok 4.20High1950#301
7Gemini 3 FlashHigh1944#175
8Claude Sonnet 4.6High1932#173
9Kimi K2.51925#266
10GPT-5.4 MiniHigh1922#406
11DeepSeek V3.21826#419
12GLM-51781#269
13Step 3.5 Flash1624#526
14MiniMax-M2.71622#539
15Grok 4.1 FastHigh1500#427

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=mathduels-solver-rating · How It Works · Data refreshed daily, snapshot 2026-11-21.